The Acoustic Guitar Forum

Go Back   The Acoustic Guitar Forum > General Acoustic Guitar and Amplification Discussion > RECORD

Reply
 
Thread Tools
  #1  
Old 09-14-2016, 07:12 PM
BoneDigger's Avatar
BoneDigger BoneDigger is offline
Moderator
 
Join Date: Jun 2012
Location: Tyler, TX
Posts: 5,046
Default Compression, limiting, and noise gates?

I am finally getting around to trying these out on my recordings. The problem is I really don't understand how they work and how the settings effect the recording. I tried compression and limiting today and it certainly made my recording louder. Not sure how to tweak these to get better recordings though?
__________________


www.mcmakinmusic.com
Reply With Quote
  #2  
Old 09-14-2016, 09:36 PM
rick-slo's Avatar
rick-slo rick-slo is offline
Charter Member
 
Join Date: Nov 2004
Location: San Luis Obispo, CA
Posts: 12,712
Default

There is tons of stuff on the internet that goes into great detail about all of this. I will just say what I do, and the general reasons why.

Personally I don't use any of the three for recording an acoustic guitar. Probably would not suggest it for voice plus guitar either.

Compression decreases the dynamic range which sucks some of the vitality out of the recording. It does however allow you to bring up the volume of the quieter passages and that can be useful say when listening in an environment with background noise (driving in your car), or listening on a low fidelity playback system.

A limiter cuts off the higher volume sounds in a more brick wall fashion (like compression but with a high compression ratio and a fast response). Basically used to prevent overs. Try to avoid overs at recording time.

If you have some very low volume (but audible) problem background noise at a specific frequency(s) you might use a frequency specifying noise gate, but otherwise they can create an audible sound artifact jungle.
__________________
Derek Coombs
Website -> Music -> Tabs -> CDs and Youtube
Guitars by Mark Blanchard, Albert&Mueller, Paul Woolson, Collings, Composite Acoustics, and Derek Coombs

Last edited by rick-slo; 09-14-2016 at 09:42 PM.
Reply With Quote
  #3  
Old 09-15-2016, 04:45 AM
ukejon ukejon is offline
Registered User
 
Join Date: Jan 2008
Location: Milwaukee
Posts: 6,483
Default

Quote:
Originally Posted by BoneDigger View Post
I am finally getting around to trying these out on my recordings. The problem is I really don't understand how they work and how the settings effect the recording. I tried compression and limiting today and it certainly made my recording louder. Not sure how to tweak these to get better recordings though?
Great advice from ric-slo. One thing to listen for is whether the compression and limiting add any unwanted non-musical background noise to your track in the process of making it louder.
__________________
My YouTube Page:
http://www.youtube.com/user/ukejon

2009 Pono koa parlor (NAMM prototype)
2014 Pono N30 DC EIR/Spruce crossover
2014 Hatcher Greta 13 fret cutaway in EIR/cedar
2017 Hatcher Josie fan fret mahogany
1973 Sigma GCR7 (OM model) rosewood and spruce
2014 Rainsong OM1000N2

....and about 5 really nice tenor ukuleles at any given moment
Reply With Quote
  #4  
Old 09-15-2016, 06:21 AM
Hot Vibrato Hot Vibrato is offline
Registered User
 
Join Date: Feb 2016
Posts: 458
Default

I don't think there are too many uses for noise gates these days for recording. There are better ways of removing unwanted noise. But compression is extremely useful for reducing peaks and for fattening up a guitar or vocal track (or drums, or whatever).

The best way to get the feel for using a compressor is to just experiment with it. Usually, if used properly, compression is subtle, but if you over-use it while experimenting, you'll get a better idea of what does what.

Typically, a lower ratio is sufficient on a vocal or guitar. Something like 4:1 is a good starting point, or even lower, and set the threshold so that it just reduces the peaks. Gain reduction of 2-3 db is plenty. A faster attack time will be appropriate for a vocal, guitar, or snare drum, while a slower attack time works better for kick drum or bass. Sometimes it helps to increase the ratio or lower the threshold so you can hear the attack and release time more clearly, and then back it off once it's dialed in. Good luck!
Reply With Quote
  #5  
Old 09-15-2016, 07:48 AM
stevecuss stevecuss is offline
Charter Member
 
Join Date: Nov 2007
Location: Front Range, Colorado
Posts: 762
Default

Quote:
Originally Posted by Hot Vibrato View Post
I don't think there are too many uses for noise gates these days for recording. There are better ways of removing unwanted noise. But compression is extremely useful for reducing peaks and for fattening up a guitar or vocal track (or drums, or whatever).

The best way to get the feel for using a compressor is to just experiment with it. Usually, if used properly, compression is subtle, but if you over-use it while experimenting, you'll get a better idea of what does what.

Typically, a lower ratio is sufficient on a vocal or guitar. Something like 4:1 is a good starting point, or even lower, and set the threshold so that it just reduces the peaks. Gain reduction of 2-3 db is plenty. A faster attack time will be appropriate for a vocal, guitar, or snare drum, while a slower attack time works better for kick drum or bass. Sometimes it helps to increase the ratio or lower the threshold so you can hear the attack and release time more clearly, and then back it off once it's dialed in. Good luck!
You've gotten great advice so far - I'd follow these settings above and you could also google "compression settings acoustic guitar recording" or something similar.

I'll +1 on the concept that compression should be subtle and you may not need it at all if you prepped well before recording. (acoustic panels to manage the room, good gain levels on the mic and good distance fro mic to guitar.) I think the way you set up your recording environment does more to help a quality recording than plugins do. I also can't believe what a difference acoustic treatment makes in a room, if you've not already done that.

Finally, if adding all those plugins (comp, limiter, gate) made your recording louder, go back in and adjust each of their gains. the only one with gain should be comp and even then, just a little gain if you are playing quietly and no gain at all if you're not.

Hope that helps.
Reply With Quote
  #6  
Old 09-15-2016, 09:07 AM
KevWind's Avatar
KevWind KevWind is offline
Moderator
 
Join Date: Apr 2008
Location: Edge of Wilderness Wyoming
Posts: 11,331
Default

Certainly some good thoughts so far.

First we need to make some clear distinctions

One distinction is when making a statement like "trying these out on my recordings." in the interest of clear communication and efficient responses . It is advisable to make the clear distinction between trying it out while "tracking" or "mixing" or both, which are actually two very distinct processes.

So the question is were you referring to using compression while actually tracking the recording, or while mixing the already tracked recording ?

Also it probably should also be noted in the audio world today ( with the millions of home recordists) that dynamics processing is likely the most misunderstood and misused FX .

Compression:
The truth is compression is a two edged sword it can be misused/overused and indeed as Rick suggested "suck the vitality" out of the music. BUT it can also in fact (if used judiciously) actually give more "vitality" (however one interprets that to mean) through increased presence and intimacy.

Also another common misconception is that a compressor is used only to "make things louder", while this "can" be a byproduct of using a compressor, it is technically not what a downward compressor actually does, which is reduction of the signal level . Only when enough make up "gain" is applied to raise the output level above the input level can compressor actually begin to raise the overall level or "make things louder".


The settings:
To clarify a downward compressor ( the most common design used) reduces the level (often called volume), of the highest peaks.(those above a certain incoming signal level called the "threshold setting ") , by an exponential mathematical amount of db's in relationship to incoming signal level.
This exponent amount, is called the "ratio" setting and is expressed as a ratio, usually like 2:1 or 4:1 etc. Which means with a 2:1 ratio, for every 1 db the level is over the "threshold" it will be reduced by 2 db exponentially . So if it is over by say 4 db it will be reduced by 8 db. (where as a 4:1 ratio will reduce that 4 db over the threshold level, by 16 db.)

"Attack" is simply how fast or quickly, the compressor will reach the full ratio setting amount of reduction.

"Release" is how long it takes the compressed signal to return to the uncompressed level.

Gain: is used to raise the output level
__________________
" Peace cannot be kept by force. It can only be achieved by understanding." Albert Einstein
Enjoy the Journey.... Kev...


KevWind at Soundcloud

Last edited by KevWind; 09-15-2016 at 10:07 AM.
Reply With Quote
  #7  
Old 09-15-2016, 01:51 PM
Mbroady's Avatar
Mbroady Mbroady is offline
Charter Member
 
Join Date: May 2008
Location: Asheville via NYC
Posts: 4,037
Default

Quote:
Originally Posted by rick-slo View Post
Personally I don't use any of the three for recording an acoustic guitar. Probably would not suggest it for voice plus guitar .
I agree with this 110%. If you have multiple tracks with multiple instruments or voices it can help In the mix down so everything sits well in the track. A little does a lot.
__________________
For Sale/trade - Custom Furch D25-SR

MJ Franks OMs, Furch Vintage 3 D-SR, Furch D32-LM,
Voyage Air VAD-2, Rainsong H-WS1000N2T, Tacoma Papoose, 72 Fender Stat
Reply With Quote
  #8  
Old 09-16-2016, 10:09 AM
BoneDigger's Avatar
BoneDigger BoneDigger is offline
Moderator
 
Join Date: Jun 2012
Location: Tyler, TX
Posts: 5,046
Default

Here is where my lack of knowledge comes in. In my DAW (Mixcraft 7), usually I record my guitar and vocals as a single recording, but with each mic (x3) on a separate input/line. I use a LDC for vocals and two SDC in an XY pattern for guitar.

I will check the input levels, then record. Once the recording is done, and it usually takes a few tries, I'll adjust the recording EQ and cut the ends to the correct spot. Then, occasionally I'll add another instrument ot two.

Once these are added, I adjust the levels and panning. I'll then go to the main control and try to adjust it to as loud as possible on the slider without clipping. I'll then save it down to MP3. The last step is that I bring the saved MP3 into a new session and adjust the volume up again to just before clipping and save it down again. That's all.

I just ordered Izotope Ozone 7, which is supposed to help with production. So, hopefully it'll help.
__________________


www.mcmakinmusic.com
Reply With Quote
  #9  
Old 09-16-2016, 04:58 PM
BoneDigger's Avatar
BoneDigger BoneDigger is offline
Moderator
 
Join Date: Jun 2012
Location: Tyler, TX
Posts: 5,046
Default

Here is a raw recording I just made for an original song I wrote. There is no processing to this, just basically record and convert to MP3. Just me and a Guild.

https://drive.google.com/file/d/0B63...w?usp=drivesdk
__________________


www.mcmakinmusic.com
Reply With Quote
  #10  
Old 09-16-2016, 08:17 PM
rick-slo's Avatar
rick-slo rick-slo is offline
Charter Member
 
Join Date: Nov 2004
Location: San Luis Obispo, CA
Posts: 12,712
Default

Quote:
Originally Posted by BoneDigger View Post
Here is where my lack of knowledge comes in. In my DAW (Mixcraft 7), usually I record my guitar and vocals as a single recording, but with each mic (x3) on a separate input/line. I use a LDC for vocals and two SDC in an XY pattern for guitar.

I will check the input levels, then record. Once the recording is done, and it usually takes a few tries, I'll adjust the recording EQ and cut the ends to the correct spot. Then, occasionally I'll add another instrument ot two.

Once these are added, I adjust the levels and panning. I'll then go to the main control and try to adjust it to as loud as possible on the slider without clipping. I'll then save it down to MP3. The last step is that I bring the saved MP3 into a new session and adjust the volume up again to just before clipping and save it down again. That's all.

I just ordered Izotope Ozone 7, which is supposed to help with production. So, hopefully it'll help.
When you have multiple inputs (instruments, voice(s)) in your recording you need to consider how to allow each one to have its place (location, frequency range, dynamic range) in the mix. Then you can use things like equalization and compression. For example equalization to prevent conflicts between a bass guitar and a regular guitar, or compression on the high dynamic range of the drums in order to be a little more in line with the other instruments.
__________________
Derek Coombs
Website -> Music -> Tabs -> CDs and Youtube
Guitars by Mark Blanchard, Albert&Mueller, Paul Woolson, Collings, Composite Acoustics, and Derek Coombs

Last edited by rick-slo; 09-17-2016 at 07:19 AM.
Reply With Quote
  #11  
Old 09-17-2016, 10:03 AM
KevWind's Avatar
KevWind KevWind is offline
Moderator
 
Join Date: Apr 2008
Location: Edge of Wilderness Wyoming
Posts: 11,331
Default

While EQ and Compression are in fact vast subjects in and of themselves, and are something that are life long learning curves (for all recording and mix engineers , not just novices ) the basic concepts are IMO integral to understanding and improving ones personal home recording pursuits and arguably critical for any multi instrument projects.

I think a common mistake (I know I was guilty of it) among those starting out, is the notion that to get the fullest sound (the most like the recordings we admire) that we must have all the sonics possible form every instrument we are using. While this may work fine for a solo instrument, as soon as we add a second instrument we are introducing possible depth and detail sucking distortion from the buildup of overlapping competing frequencies, and becomes more important as we add instruments. Even just acoustic guitar and the human voice are a primary examples of how this can apply. Given they both occupy a significant portion of the same frequency range.


To add to what Rick has stated above
What I do as always the first, and often only EQ'ing I use when mixing multiple tracks and instruments, is to do exactly what (my take is on what) he is describing, to carve out some space for each instrument so as to have each be a bit more detailed by being differentiated from one another .

So here is just a brief summery of how I approach EQ.

I start this process by using "subtractive eq".
It should be noted that in multi instrument tracks (even just guitar and voice) but even more so as you add instruments. Sonic buildup of conflicting overlapping tones tends to act as masking and possible distortion. Masking and distortion always results in less perceived clarity which tends to make recordings seem to have less depth.

The first subtractive EQ I do, is roll off the low end (steeply) on every track ( it of course takes experimenting on exactly where and how much) but usually somewhere between 50 and 150 hz
I will often for example roll off my vocal at 80 and my guitar at 100 or 125

BoneDigger, I checked the Mixcraft web sight it looks like the free version has only a graphic EQ which will work but is not as good as the TB Parametric EQ available in the Pro version (but is actually also downloadable as an additional EQ )

The nest step I do is what Ronan Chris Murphy calls "seek and destroy", this step is by far easier and more accurate with a multi band parametric eq with adjustable Q like TB Parametric,
On each track start with a very narrow Q (band width) I then boost the level (towards the upper limit) and sweep the frequency range and find what frequencies are sounding boxy honky (somewhat like hearing the sound thru a large metal culvert.) Once it find that frequency where that honk is the most pronounced I simply cut somewhere between -2 and -6 db in a fairly narrow but wider Q than I used for sweeping.

For guitar & vocal I find the most pronounced honk frequencies in each and if they different by enough say 30 hz or more I cut at those freqs'. If they are very close then I fudge the cut point up on one, and down on the other (so I am cutting different fairly narrow ranges) on each . This subtractive EQ'ing is often enough EQ to get the the guitar and vocal to stand out a bit more from each other .

Again I always do subtractive (cutting) before ever doing any additive (boosting) EQ and often additive is actually unnecessary.


Again just to make a point low end build up is probably the single most egregious in terms of having the effect of flatting and sucking the depth out of a recording. So much so that for example (even though it seem counter intuitive) with bass guitar and kick drum the worst thing you can do IMO is boost (without having first cut) either one or both with EQ, even in different ranges. If you want to have the bass and kick more distinct from each other start by cutting first. If you listen closely to raw tracks of kick and bass it is actually pretty amazing that by simply rolling off the low end on one will make both more distinct. And rolling off the low end on both but in different starting frequencies, they will begin to #1 be even more distinct and #2 work more complimentary with each other instead of fighting each other for attention.
__________________
" Peace cannot be kept by force. It can only be achieved by understanding." Albert Einstein
Enjoy the Journey.... Kev...


KevWind at Soundcloud

Last edited by KevWind; 09-17-2016 at 10:14 AM.
Reply With Quote
  #12  
Old 09-17-2016, 12:01 PM
BoneDigger's Avatar
BoneDigger BoneDigger is offline
Moderator
 
Join Date: Jun 2012
Location: Tyler, TX
Posts: 5,046
Default

I do indeed have the Pro Version of Mixcraft. Unfortunately, I have been having a hard time getting it to accept my Focusrite ASIO drivers. The recording I made above was in Reaper, which accepts the ASIO just fine.

As a raw recording, can you tell me what you would do to the track I posted above?
__________________


www.mcmakinmusic.com
Reply With Quote
  #13  
Old 09-17-2016, 12:37 PM
KevWind's Avatar
KevWind KevWind is offline
Moderator
 
Join Date: Apr 2008
Location: Edge of Wilderness Wyoming
Posts: 11,331
Default

Quote:
Originally Posted by BoneDigger View Post
I do indeed have the Pro Version of Mixcraft. Unfortunately, I have been having a hard time getting it to accept my Focusrite ASIO drivers. The recording I made above was in Reaper, which accepts the ASIO just fine.

As a raw recording, can you tell me what you would do to the track I posted above?
Unfortunately I am not at my studio and wont be until Oct 6
so all I have to listen with is ear buds on my laptop
but the first thing I would do is roll off the low end of the guitar. You will need to experiment but start with 60 hz and slowly go up from there until you really notice the low end of the Git going away then go back down a little. Also try it soloed and with the vocal and see it things become just a little clearer .
I have to go so I'll perhaps post some more when I get back.
__________________
" Peace cannot be kept by force. It can only be achieved by understanding." Albert Einstein
Enjoy the Journey.... Kev...


KevWind at Soundcloud
Reply With Quote
  #14  
Old 09-17-2016, 03:19 PM
BoneDigger's Avatar
BoneDigger BoneDigger is offline
Moderator
 
Join Date: Jun 2012
Location: Tyler, TX
Posts: 5,046
Default

See if this sounds any better.

https://drive.google.com/file/d/0B63...w?usp=drivesdk
__________________


www.mcmakinmusic.com
Reply With Quote
  #15  
Old 09-18-2016, 08:25 AM
KevWind's Avatar
KevWind KevWind is offline
Moderator
 
Join Date: Apr 2008
Location: Edge of Wilderness Wyoming
Posts: 11,331
Default

Quote:
Originally Posted by BoneDigger View Post
As I said I hesitate to make any mix judgements calls because I am not able to monitor on an accurate system.

First off (as I was trying to explain in another current thread) the significant level boost alone makes any accurate assessment of "actual " improvement vs. perceived improvement invalid. And honestly that is best determined on your system while you are mixing it by making a one slight adjustment listen, then immediately bypass and listen

With that said, and understanding that ear buds are a highly questionable tool ,

I am guessing you did more than slightly roll off the very low end of the guitar??

Perhaps you can post some screen shots or a laundry list of what the actual mix moves and final changes you made for the new version vs the RAW version

So I what I did was click on both links so that they each had a tab in my browser so I could quickly go back and forth. I then tried to get the levels more matched between the RAW tracks and the new ones so I could try to get a more accurate picture.

First it sounds like the low end boom/mush is greatly improved and I think there is better overall detail . I am curious how, where, and how much you cut the lows? i.e. the laundry list
But it also sounds like the mids and or upper mids are somewhat exaggerated, if not boosted by eq then because of "equal loudness contour " when when boosted by level becomes perceived to be more pronounced

Also your voice has a good bit of dynamic change and a certain grit in the mid range and is fairly pronounced in that range (which BTW is a good thing and helps make it more unique and by no means am I suggesting you change your singing style ) But that mid grit especially when you are pushing it dynamically can also can be easily become a bit to pronounced and almost sound like the beginning of distortion. And it competes directly with and can overpower that range in your guitar.
Did you do any of the search, locate and cut technique I spoke of ?


If I were recording and mixing this song for you . I would use my standard procedure (and what all my templates reflect) all audio tracks would have a multi-band parametric EQ (in the first plug in position) and Compressor (in the second plugin position)

For tracking/recording the EQ would be bypassed but on your vocal (depending on the dynamic swing I was seeing in the levels) I might record with the comp engaged on your vocal, set to only a 2:1 ratio, a medium to slow attack, medium or slightly quicker release and no gain.
Or if I was given these raw tracks and was only mixing. On the vocal track I would have a Comp in the first position, an EQ second and a second Comp in the third position. I would set the first comp like above as tracking comp. Then I would search and slightly cut the most pronounced grit with EQ, then depending on how that sounded like, I might use the second Comp with something like a 2:1 or 4:1 ratio, slow attack and medium release

On your guitar I would record dry but when mixing would find frequency the most "Honk" and cut slightly - 2 to -4 db
I would also consider using a slight bit of reverb in parallel and send both the vocal and git tracks
__________________
" Peace cannot be kept by force. It can only be achieved by understanding." Albert Einstein
Enjoy the Journey.... Kev...


KevWind at Soundcloud

Last edited by KevWind; 09-18-2016 at 09:01 AM.
Reply With Quote
Reply

  The Acoustic Guitar Forum > General Acoustic Guitar and Amplification Discussion > RECORD

Thread Tools



All times are GMT -6. The time now is 02:51 AM.


Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2019, vBulletin Solutions, Inc.
Copyright ©2000 - 2018, The Acoustic Guitar Forum
vB Ad Management by =RedTyger=